Associations of urinary sodium excretion with central hemodynamics and changes in vascular structure and function at high altitude

Abstract Research reports on associations of urinary sodium excretion with central hemodynamic parameters and vascular changes are quite limited in general or non‐hypertensive population. The purpose of the current study was to explore such associations in Chinese general Tibetans living at high altitude. This cross‐sectional study was conducted in Luhuo County, Ganzi Tibetan Autonomous Prefecture with average elevation of 3800 meters from December 2018 to January 2019. A total of 294 Tibetans were included in the current study. Twenty‐four hour urinary sodium excretion was estimated by second fasting spot urine in the morning using Kawasaki formula. Central hemodynamic parameters, including central systolic blood pressure (CSBP), central diastolic blood pressure (CDBP), central pulse pressure (CPP), central mean arterial pressure (CMAP), augmentation pressure (AP), and augmentation index standardized for heart rate of 75 (AIx75), were evaluated using the SphygmoCor system. Vascular structures and functions were assessed by carotid intima media thickness (CIMT) test and brachial ankle pulse wave velocity (baPWV), respectively. Estimated mean 24h urinary sodium excretion of Tibetans in Luhuo County was 5.26±1.61 g. After adjustment, estimated 24h urinary sodium was positively associated with CSBP (β = 1.15, p = .008) and CPP (β = 0.87, p = .013). Line graph of means across urinary sodium quartiles showed that associations of 24 h urinary sodium excretion with AIx75 and baPWV presented approximate “J” shape after controlling for confounders. Estimated 24 h sodium excretion was independently and positively associated with CSBP and CPP. Moreover, association between urinary sodium excretion and arterial elasticity, as evaluated by baPWV and AIx75, presented “J” shape. Further studies are needed to verify J‐shaped association and “safe” zone of sodium intake.


INTRODUCTION
Previous studies reported that high-salt diet is among the most important contributing factors to pathogenesis of hypertension and leads to vascular remodeling and arteriosclerosis through inflammatory responses, oxidative stress, and activation of renin-angiotensinaldosterone system (RAAS). 1 Remodeling of central and peripheral arteries increases amplitude of forward and reflected waves resulting in elevation of central blood pressure. Central arterial pressure represents direct pressure load exposed to left ventricle and has a substantial impact on perfusion of coronary artery and brain. Previous studies revealed that central arterial pressure is more closely correlated with vascular diseases, target organ damage of hypertension, and cardiovascular (CV) events compared to brachial artery pressure. 2,3 There has been little focus on associations of salt intake with cen-

Study design and population
This cross-sectional study was carried out in Luhuo County, Ganzi

BaPWV measurement
BaPWV is a noninvasive method for detecting arterial stiffness, which has been validated with carotid femoral pulse wave velocity (cfPWV) 6 and invasive PWV. 7 BaPWV has been used widely in prospective studies for prediction of CV diseases 8 and all-cause mortality, 9 mainly in Asia.
BaPWV examination was conducted using Omron oscillometrybased device (VP1000 BP-203RPE-III, ColinCo, Ltd, Komaki, Japan) in warm and quiet room. After 5-min rest in supine position, four cuffs were wrapped in bilateral upper arms and ankles of patients, two ECG pads were clipped to both wrists and one heart sound sensor was placed at the corner of sternum. Pulse wave velocity was computed automatically based on estimated distances and time consumed between sampling points. Average values of left and right sides were used for analysis.

Statistical analysis
Quantitative data were presented as mean ± SD or median (P 25 , P 75 ) and qualitative data were presented as percentages or frequencies.
Continuous variables in different groups were compared using analysis of variance. Differences in categorical variables among groups were compared using χ 2 tests. Partial correlation analysis was used to ana-

Associations of urinary sodium excretion with peripheral and central hemodynamics
Partial correlation analysis in the current study showed that estimated 24 h urinary sodium excretion was positively correlated with office SBP (r = 0.178, p = .005) and office PP (r = 0.195, p = .002) but it was not correlated with office DBP (r = 0.029, p = .651) and office MAP (r = 0.099, p = .122).
In addition, partial correlation analysis showed that estimated 24 h urinary sodium excretion was positively correlated with CSBP and CPP (r = 0.176, p = .006 and r = 0.176, p = .011, respectively). However, correlations of estimated 24h urinary sodium excretion with CDBP, CMAP, and Ln AP (p = .380, .069, and .465, respectively) were not significant. Estimated 24h urinary sodium excretion was significantly and negatively correlated with AIx 75 (p = .036). These findings are presented in Table 2. Multiple linear regression analysis showed that 24 h urinary sodium excretion was independently and positively associated with CSBP and CPP (β = 1.15, p = .008 and β = 0.87, p = .013, respectively). However, associations of estimated 24 h urinary sodium excretion with CDBP, CMAP, and Ln AP were not significant. Findings of the current study also established that 24 h urinary sodium excretion was significantly and negatively associated with AI X75 (β = -1.12, p = .036) ( Table 3).

Associations of urinary sodium excretion with baPWV and CIMT
Partial correlation analysis showed that estimated 24 h urinary sodium excretion was negatively correlated with baPWV (p = .005) but it was not correlated with CIMT (p = .168) ( Table 2).
After controlling for confounders, line graph of means across 24 h urinary sodium excretion quartiles showed that there was a positive linear association between 24 h urinary sodium excretion with CSBP and CPP. Associations of 24 h urinary sodium excretion with baPWV and AIx 75 presented "J" shape as shown in Figure 1.

DISCUSSION
Findings of the current study established that estimated 24 h urinary sodium excretion of native general Tibetans living in Luhuo County was 5.26±1.61 g and estimated 24 h salt intake was 13.36±4.09 g.
In addition, the current study showed that estimated 24 h urinary sodium excretion was independently and positively associated with  The study also showed that association between high urinary sodium excretion and CV events was only significant in patients with hypertension and such relevance weakened when blood pressure was adjusted, indicating that high salt intake contributed to CV diseases by elevating blood pressure. Nevertheless, in low sodium excretion group, the association was evident in patients with or without hypertension and remained significant when controlling for blood pressure, implying that low salt intake led to high risk of CV diseases independent of blood pressure. To our knowledge, arterial stiffness is closely related to CV diseases. Therefore, exploratory studies on relationship between urinary sodium excretion with arterial stiffness could provide clues for association of salt intake with CV events. urinary sodium excretion of 3.6 g. They speculated that J-shaped association was not observed due to low urinary sodium excretion.
Another study 18  Based on findings of the current study, dietary salt restriction is essential for control of central blood pressure considering current status of high salt intake in this high-altitude area. J-shaped association between urinary sodium with vascular changes should be confirmed in future studies by enlarging sample sizes from more sites in high altitude Tibetans area and collecting 24 h urine samples. A recent study revealed that daily pattern of urinary sodium excretion, aside from the amount of sodium excretion, was associated with central blood pressure and arterial stiffness. 26 This implies that intraindividual pattern of urinary sodium excretion should also be explored in future studies to improve individualized risk stratification and management of hypertension and other CV diseases.

ACKNOWLEDGMENTS
The authors acknowledge the staffs and participants of the CDC of Luhuo County, Ganzi Tibetan autonomous prefecture and Luhuo People's Hospital. We also acknowledge the members of our team for our efforts and cooperation to complete the research under the cold and hypoxic environment. We also thank the Home for Researchers editorial team (www.home-for-researchers.com) for editing the language of the manuscript. This study was funded by Science and Technology Pillar Programs in Sichuan Province (Grant No. 2017SZ0008). The funding body had no role in the design of the study, collection, analysis and interpretation of the data, or writing the manuscript.

CONFLICT OF INTERESTS
On behalf of all authors, the corresponding author states that there is no conflict of interest.